Added yolov8 model (#2)

- Added yolov8 model (9c5ce294b0d813a7094b44b2d4d0635726727d53)
- Update app.py (3f98b8c0224f0fbed5840b9827314bf5dbefae08)
- Update requirements.txt (d139267820aeca22b668bf32e8f7eb556a1c9910)


Co-authored-by: Kadir Nar <[email protected]>

README.md

@@ -6,7 +6,7 @@ colorTo: yellow
 sdk: gradio
 app_file: app.py
 pinned: false
-duplicated_from: fcakyon/sahi-yolov5
+duplicated_from: deprem-ml/deprem_satellite_test
 license: openrail
 ---
 

app.py

@@ -1,34 +1,21 @@
-import gradio as gr
-import sahi.utils
-from sahi import AutoDetectionModel
-import sahi.predict
-import sahi.slicing
+from sahi import utils, predict, AutoDetectionModel
 from PIL import Image
+import gradio as gr
 import numpy
-from huggingface_hub import hf_hub_download
 import torch
+import os
 
+os.system('pip install git+https://github.com/fcakyon/ultralyticsplus.git')
 
-IMAGE_SIZE = 640
-
-model_id = 'deprem-ml/Binafarktespit-yolo5x-v1-xview'
-
-
+model_id_list = ['deprem-ml/Binafarktespit-yolo5x-v1-xview', 'SerdarHelli/deprem_satellite_labeled_yolov8']
 current_device = "cuda" if torch.cuda.is_available() else "cpu"
-model_types = ["YOLOv5", "YOLOv5 + SAHI"]
-# Model
-model = AutoDetectionModel.from_pretrained(
-    model_type="yolov5",
-    model_path=model_id,
-    device=current_device,
-    confidence_threshold=0.5,
-    image_size=IMAGE_SIZE,
-)
-
+model_types = ["YOLOv5", "YOLOv5 + SAHI", "YOLOv8"]
 
-def sahi_yolo_inference(
-    model_type,
+def sahi_yolov5_inference(
     image,
+    model_id,
+    model_type,
+    image_size,
     slice_height=512,
     slice_width=512,
     overlap_height_ratio=0.1,
@@ -39,26 +26,43 @@ def sahi_yolo_inference(
     postprocess_class_agnostic=False,
 ):
 
-    # image_width, image_height = image.size
-    # sliced_bboxes = sahi.slicing.get_slice_bboxes(
-    #     image_height,
-    #     image_width,
-    #     slice_height,
-    #     slice_width,
-    #     False,
-    #     overlap_height_ratio,
-    #     overlap_width_ratio,
-    # )
-    # if len(sliced_bboxes) > 60:
-    #     raise ValueError(
-    #         f"{len(sliced_bboxes)} slices are too much for huggingface spaces, try smaller slice size."
-    #     )
-
     rect_th = None or max(round(sum(image.size) / 2 * 0.0001), 1)
     text_th = None or max(rect_th - 2, 1)
+    
+    if model_type == "YOLOv5":
+        # standard inference
+        model = AutoDetectionModel.from_pretrained(
+            model_type="yolov5",
+            model_path=model_id,
+            device=current_device,
+            confidence_threshold=0.5,
+            image_size=image_size,
+            )
+        
+        prediction_result_1 = predict.get_prediction(
+            image=image, detection_model=model
+        )
 
-    if "SAHI" in model_type:
-        prediction_result_2 = sahi.predict.get_sliced_prediction(
+        visual_result_1 = utils.cv.visualize_object_predictions(
+            image=numpy.array(image),
+            object_prediction_list=prediction_result_1.object_prediction_list,
+            rect_th=rect_th,
+            text_th=text_th,
+        )
+        
+        output = Image.fromarray(visual_result_1["image"])
+        return output
+    
+    elif model_type == "YOLOv5 + SAHI":
+        model = AutoDetectionModel.from_pretrained(
+            model_type="yolov5",
+            model_path=model_id,
+            device=current_device,
+            confidence_threshold=0.5,
+            image_size=image_size,
+        )
+        
+        prediction_result_2 = predict.get_sliced_prediction(
             image=image,
             detection_model=model,
             slice_height=int(slice_height),
@@ -70,54 +74,38 @@ def sahi_yolo_inference(
             postprocess_match_threshold=postprocess_match_threshold,
             postprocess_class_agnostic=postprocess_class_agnostic,
         )
-        visual_result_2 = sahi.utils.cv.visualize_object_predictions(
+        
+        visual_result_2 = utils.cv.visualize_object_predictions(
             image=numpy.array(image),
             object_prediction_list=prediction_result_2.object_prediction_list,
             rect_th=rect_th,
             text_th=text_th,
         )
+        
         output = Image.fromarray(visual_result_2["image"])
         return output
 
-    else:
-        # standard inference
-        prediction_result_1 = sahi.predict.get_prediction(
-            image=image, detection_model=model
-        )
-        print(image)
-        visual_result_1 = sahi.utils.cv.visualize_object_predictions(
-            image=numpy.array(image),
-            object_prediction_list=prediction_result_1.object_prediction_list,
-            rect_th=rect_th,
-            text_th=text_th,
-        )
-        output = Image.fromarray(visual_result_1["image"])
-        return output
-
-    # sliced inference
+    elif model_type == "YOLOv8":
+        from ultralyticsplus import YOLO, render_result
 
+        model = YOLO('SerdarHelli/deprem_satellite_labeled_yolov8')
+        result = model.predict(image, imgsz=image_size)[0]
+        render = render_result(model=model, image=image, result=result, rect_th=rect_th, text_th=text_th)
+        return render
 
 inputs = [
-    gr.Dropdown(
-        choices=model_types,
-        label="Choose Model Type",
-        type="value",
-        value=model_types[1],
-    ),
     gr.Image(type="pil", label="Original Image"),
-    gr.Number(default=512, label="slice_height"),
-    gr.Number(default=512, label="slice_width"),
-    gr.Number(default=0.1, label="overlap_height_ratio"),
-    gr.Number(default=0.1, label="overlap_width_ratio"),
-    gr.Dropdown(
-        ["NMS", "GREEDYNMM"],
-        type="value",
-        value="NMS",
-        label="postprocess_type",
-    ),
-    gr.Dropdown(["IOU", "IOS"], type="value", value="IOU", label="postprocess_type"),
-    gr.Number(value=0.5, label="postprocess_match_threshold"),
-    gr.Checkbox(value=True, label="postprocess_class_agnostic"),
+    gr.Dropdown(choices=model_id_list,label="Choose Model",value=model_id_list[0]),
+    gr.Dropdown( choices=model_types, label="Choose Model Type", value=model_types[1]),
+    gr.Slider(minimum=128, maximum=2048, value=640, step=32, label="Image Size"),
+    gr.Slider(minimum=128, maximum=2048, value=512, step=32, label="Slice Height"),
+    gr.Slider(minimum=128, maximum=2048, value=512, step=32, label="Slice Width"),
+    gr.Slider(minimum=0.0, maximum=1.0, value=0.1, step=0.1, label="Overlap Height Ratio"),
+    gr.Slider(minimum=0.0, maximum=1.0, value=0.1, step=0.1, label="Overlap Width Ratio"),
+    gr.Dropdown(["NMS", "GREEDYNMM"], type="value", value="NMS", label="Postprocess Type"),
+    gr.Dropdown(["IOU", "IOS"], type="value", value="IOU", label="Postprocess Type"),
+    gr.Slider(minimum=0.0, maximum=1.0, value=0.5, step=0.1, label="Postprocess Match Threshold"),
+    gr.Checkbox(value=True, label="Postprocess Class Agnostic"),
 ]
 
 outputs = [gr.outputs.Image(type="pil", label="Output")]
@@ -126,13 +114,14 @@ title = "Building Detection from Satellite Images with SAHI + YOLOv5"
 description = "SAHI + YOLOv5 demo for building detection from satellite images. Upload an image or click an example image to use."
 article = "<p style='text-align: center'>SAHI is a lightweight vision library for performing large scale object detection/ instance segmentation.. <a href='https://github.com/obss/sahi'>SAHI Github</a> | <a href='https://medium.com/codable/sahi-a-vision-library-for-performing-sliced-inference-on-large-images-small-objects-c8b086af3b80'>SAHI Blog</a> | <a href='https://github.com/fcakyon/yolov5-pip'>YOLOv5 Github</a> </p>"
 examples = [
-    [model_types[1], "26.jpg", 512, 512, 0.1, 0.1, "NMS", "IOU", 0.25, False],
-    [model_types[1], "27.jpg", 512, 512, 0.1, 0.1, "NMS", "IOU", 0.25, False],
-    [model_types[1], "28.jpg", 512, 512, 0.1, 0.1, "NMS", "IOU", 0.25, False],
-    [model_types[1], "31.jpg", 512, 512, 0.1, 0.1, "NMS", "IOU", 0.25, False],
+    ["data/26.jpg", 'deprem-ml/Binafarktespit-yolo5x-v1-xview', "YOLOv5 + SAHI", 640, 512, 512, 0.1, 0.1, "NMS", "IOU", 0.25, False],
+    ["data/27.jpg", 'deprem-ml/Binafarktespit-yolo5x-v1-xview', "YOLOv5 + SAHI", 640, 512, 512, 0.1, 0.1, "NMS", "IOU", 0.25, False],
+    ["data/28.jpg", 'deprem-ml/Binafarktespit-yolo5x-v1-xview', "YOLOv5 + SAHI", 640, 512, 512, 0.1, 0.1, "NMS", "IOU", 0.25, False],
+    ["data/31.jpg", 'deprem-ml/SerdarHelli-yolov8-v1-xview', "YOLOv8", 640, 512, 512, 0.1, 0.1, "NMS", "IOU", 0.25, False],
 ]
-gr.Interface(
-    sahi_yolo_inference,
+
+demo = gr.Interface(
+    sahi_yolov5_inference,
     inputs,
     outputs,
     title=title,
@@ -141,4 +130,6 @@ gr.Interface(
     examples=examples,
     theme="huggingface",
     cache_examples=True,
-).launch(debug=True, enable_queue=True)
+)
+
+demo.launch(debug=True, enable_queue=True)

requirements.txt

@@ -1,5 +1,4 @@
-torch==1.10.2+cpu
-torchvision==0.11.3+cpu
--f https://download.pytorch.org/whl/torch_stable.html
+torch==1.10.2
+torchvision==0.11.3
 yolov5==7.0.8
 sahi==0.11.11